Post on 03-Apr-2018
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
1/236
TEXTOS DE MATEMATICA
INTRODUCTION
TO
CHARACTERISTIC CLASSESAND
INDEX THEORY
Jean-Pierre Schneiders
(X) =
#([].[]) =
XeTX
w(EF)=
w
(E)w(F)
(X;OX(D))=
degD+
1g (
X;F)
=
XchF
tdT
X
UNIVERSIDADE DE LISBOA
Faculdade de Ciencias
Departamento de Matematica
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
2/236
Textos de Matematica, Volume 13,
Departamento de MatematicaFaculdade de Ciencias da Universidade de Lisboa, 2000
Editores: Fernando C. Silva e L. Trabucho
Ttulo: Introduction to Characteristic Classes and Index Theory
Autor: Jean-Pierre Schneiders
ISBN: 972-8394-12-8
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
3/236
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
4/236
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
5/236
INTRODUCTION
TO
CHARACTERISTIC CLASSES
AND
INDEX THEORY
Jean-Pierre Schneiders
UNIVERSIDADE DE LISBOA
Faculdade de Ciencias
Departamento de Matematica
2000
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
6/236
Authors address:
Laboratoire Analyse, Geometrie et Applications (UMR 7539)
Institut Galilee, Universit e Paris 13Avenue Jean-Baptiste Clement
F-93430 Villetaneuse, France
E-mail:
jps@math.univ-paris13.fr
URL:
Mathematics Subject Classification (2000): 55R40, 57R20, 19L10, 14C40.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
7/236
Preface
This book is based on a course given by the author at the university of
Lisbon during the academic year 19971998. This course was divided in
three parts dealing respectively with characteristic classes of real and com-
plex vector bundles, Hirzebruch-Riemann-Roch formula and Atiyah-Singer
theorem. In the text which follows, we have decided to treat only the first
two subjects. For an introduction to the last one, we refer the reader to [21]
for the classical point of view or to [22, 23] for recent developments.
The theory of characteristic classes is a very well developed branch of
mathematics and the literature concerning Riemann-Roch theorem is huge.
So, we will not try to give a full view of these subjects. We will rather
present a few basic but fundamental facts which should help the reader to
gain a good idea of the mathematics involved.
Although the reader is assumed to have a good knowledge of homological
algebra and topology, we begin with a chapter surveying the results of sheaf
theory which are needed in the rest of the book. In particular we recallresults concerning acyclic sheaves, taut subspaces, Poincare-Verdier duality
and Borel-Moore homology, illustrating them by means of examples and
exercises.
We refer the reader who would like a more detailed treatment of this part
to standard texts on sheaf theory (e.g. [11, 6, 28, 18]) and algebraic topology
(e.g. [9, 29, 14]). Older works may also be of interest (e.g. [31, 25, 1, 19]).
Chapter 2 is devoted to Euler classes. As a motivation, we begin by prov-
ing the classical Lefschetz fixed point formula and applying it to computethe Euler-Poincare characteristic of a compact oriented topological mani-
fold by means of its Euler class. Next, we study Thom and Euler classes of
oriented real vector bundles. In particular, we consider Thom isomorphism,
Gysin exact sequence and functorial properties of Euler classes. We end
iii
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
8/236
iv Preface
with results on the Euler class of a normal bundle which allow us to link the
Euler class of an oriented differential manifold with the one of its tangent
bundle.The first part of Chapter 3 deals mainly with Stiefel-Whitney classes.
We define them in the Grothendieck way by means of projective bundles
and Euler classes of associated tautological bundles. Then, we establish the
pull-back and direct sum formulas. Thanks to the splitting principle, we
also obtain a formula for the Stiefel-Whitney classes of a tensor product. In
the second part of the chapter, we study in general the characteristic classes
of real vector bundles. We begin by classifying real vector bundles of rank
r by means of homotopy classes of maps with values in the infinite Grass-
mannian G,r. This establishes a link between the characteristic classesof real vector bundles with coefficients in a group M and the elements of
H(G,r; M). By computing this last group for M = Z2, we show that the
modulo 2 characteristic classes of real vector bundles are polynomials in
the Stiefel-Whitney classes. We end by explaining the usual cohomological
classification of real vector bundles and deducing from it that line bundles
are classified by their first Stiefel-Whitney class. This allows us to give a
criterion for a real vector bundle to be orientable.
Chapter 4 is centered on Chern classes. We begin by adapting most ofthe results concerning Stiefel-Whitney classes to the complex case. Next,
we consider specific results such as the Chern-Weil method of computing
Chern classes using the curvature of a connection. We also treat briefly of
the Chern character. The last part of the chapter is little bit technical. It is
devoted to Iversens construction of the local Chern character for complexes
of complex vector bundles (see [17]) and to its application to the definition
of a local Chern character for coherent analytic sheaves.
For more details on the three preceding chapters, the reader may consults
classical books on the theory of fiber bundles (e.g. [30, 20, 16]).The last chapter of this book is about Riemann-Roch theorem. After
a short review of the finiteness and duality results for coherent analytic
sheaves, we reach the central question of this part i.e. how to compute
(X; F)
for a coherent analytic sheaf on a compact complex analytic manifold X.
The answer to this question is essentially due to Hirzebruch (see [15]) and
states that(X; F) =
X
ch F td T X
where ch F is the Chern character of F and td T X is the Todd class of thetangent bundle of X. To better understand the meaning of this formula, we
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
9/236
Preface v
devote Sections 24 to the easy case of line bundles on complex curves. In
this situation, X is a compact Riemann surface and we can link the gener-
alized Riemann-Roch theorem considered above with the original results ofRiemann and Roch. We end the chapter by proving Hirzebruch-Riemann-
Roch theorem for complex projective manifolds. We follow Grothendieck
approach (see [5]) by reducing the result to the case of the complex projec-
tive space by means of a relative Riemann-Roch formula for embeddings.
However, to treat this relative case, we have not followed [5] but used a sim-
pler method based on the ideas of [2, 3, 4] and the proof of the Grothendieck-
Riemann-Roch formula in [10].
For bibliographical informations concerning the subject treated in this
chapter we refer to [15, 10]. Interesting historical comments may also befound in [7, 8].
It is a pleasure to end this preface by thanking heartily the CMAF for
its hospitality during my stay at Lisbon university. I think in particular to
T. Monteiro Fernandes who invited me to give the course on which this book
is based and suggested to publish it in this collection. I am also grateful
to her for taking a set of lectures notes which served as a first draft for
this work. All my thanks also to O. Neto and to the various people who
attended the course and whose interest has been a strong motivation forturning the raw lecture notes into a book.
Let me finally thank F. Prosmans whose help was invaluable at all the
stages of the preparation of the manuscript.
March 2000 Jean-Pierre Schneiders
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
10/236
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
11/236
Contents
1 Survey of sheaf theory 1
1.1 Abelian presheaves and sheaves . . . . . . . . . . . . . . . . . 1
1.2 Sections of an abelian sheaf . . . . . . . . . . . . . . . . . . . 7
1.3 Cohomology with supports . . . . . . . . . . . . . . . . . . . 9
1.4 Flabby and soft abelian sheaves . . . . . . . . . . . . . . . . . 11
1.5 Cohomology of subspaces and tautness . . . . . . . . . . . . . 13
1.6 Excision and Mayer-Vietoris sequences . . . . . . . . . . . . . 14
1.7 Inverse and direct images . . . . . . . . . . . . . . . . . . . . 21
1.8 Homotopy theorem . . . . . . . . . . . . . . . . . . . . . . . . 31
1.9 Cohomology of compact polyhedra . . . . . . . . . . . . . . . 40
1.10 Cohomology of locally compact spaces . . . . . . . . . . . . . 48
1.11 Poincare-Verdier duality . . . . . . . . . . . . . . . . . . . . . 51
1.12 Borel-Moore homology . . . . . . . . . . . . . . . . . . . . . . 58
1.13 Products in cohomology and homology . . . . . . . . . . . . . 60
1.14 Cohomology of topological manifolds . . . . . . . . . . . . . . 65
1.15 Sheaves of rings and modules . . . . . . . . . . . . . . . . . . 72
2 Euler class of manifolds and real vector bundles 73
2.1 Lefschetz fixed point formula . . . . . . . . . . . . . . . . . . 73
2.2 Euler classes of manifolds and index theorem . . . . . . . . . 81
2.3 Basic notions on real vector bundles . . . . . . . . . . . . . . 81
2.4 Orientation of real vector bundles . . . . . . . . . . . . . . . . 83
2.5 Thom isomorphism and Gysin exact sequence . . . . . . . . . 87
2.6 Euler classes of inverse images and direct sums . . . . . . . . 90
2.7 Euler classes of normal bundles . . . . . . . . . . . . . . . . . 94
vii
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
12/236
viii Contents
3 Characteristic classes of real vector bundles 99
3.1 Stiefel-Whitney classes . . . . . . . . . . . . . . . . . . . . . . 99
3.2 Splitting principle and consequences . . . . . . . . . . . . . . 1063.3 Homotopical classification of real vector bundles . . . . . . . 118
3.4 Characteristic classes . . . . . . . . . . . . . . . . . . . . . . . 126
3.5 Cohomological classification of real vector bundles . . . . . . 133
4 Characteristic classes of complex vector bundles 139
4.1 Generalities on complex vector bundles . . . . . . . . . . . . . 139
4.2 Chern classes . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
4.3 Chern-Weil construction . . . . . . . . . . . . . . . . . . . . . 146
4.4 Chern character . . . . . . . . . . . . . . . . . . . . . . . . . . 1564.5 Local chern character . . . . . . . . . . . . . . . . . . . . . . 161
4.6 Extension to coherent analytic sheaves . . . . . . . . . . . . . 174
5 Riemann-Roch theorem 179
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
5.2 Cohomology of compact complex curves . . . . . . . . . . . . 186
5.3 Divisors on complex curves . . . . . . . . . . . . . . . . . . . 188
5.4 Classical Riemann and Roch theorems . . . . . . . . . . . . . 190
5.5 Cohomology of coherent analytic sheaves on Pn(C) . . . . . . 1985.6 Hirzebruch-Riemann-Roch theorem for Pn(C) . . . . . . . . . 207
5.7 Riemann-Roch for holomorphic embeddings . . . . . . . . . . 208
5.8 Proof of Hirzebruch-Riemann-Roch theorem . . . . . . . . . . 214
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
13/236
1
Survey of sheaf theory
1.1 Abelian presheaves and sheaves
Let X be a topological space and let Op(X) denote the category of opensubsets of X and inclusion maps.
Definition 1.1.1. An abelian presheaf on X is a functor F : Op(X)op Ab where Ab denotes the category of abelian groups. In other words, anabelian presheaf is a law which associates an abelian group F(U) to any
open subset U of X and which associates to any open subset V U a
restriction morphism
rFV U : F(U) F(V)
in such a way that
rFWV rFV U = r
FWU
for any chain of open subsets W V U of X. We often denote rFV U(s)simply by s|V when there is no risk of confusion.
A morphism of abelian presheaves is simply a morphism of the cor-responding functors. More explicitly, a morphism of abelian presheaves
f : F G is a law which associates to any open subset U ofX a morphism
f(U) : F(U) G(U)
1
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
14/236
2 1. Survey of sheaf theory
in such a way that the diagram
F(U)f(U)
GG
rFV U
G(U)
rGV U
F(V)f(V)
GG G(V)
is commutative for any open subsets V U of X.With this notion of morphisms, abelian presheaves form an abelian cat-
egory, denoted Psh(X).
Examples 1.1.2.
(a) On X, we may consider the abelian presheaf C0,X defined by settingC0,X(U) = {f : U C : f continuous}
and defining the restriction morphisms by means of the usual restric-
tions of functions.
(b) If X is endowed with a Borelian measure , we may consider the
abelian presheaves Lp,X defined by associating to an open U of X
the quotient of the abelian group
{f : U C : U
|f|pdV +}
by the subgroup
{f : U C : f = 0 almost everywhere};
the restriction morphisms being the obvious ones.
Definition 1.1.3. An abelian sheaf on X is an abelian presheaf F suchthat
(a) we have
F() = 0;
(b) for any open coveringU of an open subset U of X, we have the exactsequence
0 F(U)U
VU
F(V)U
V,WU
F(V W)
where U(s) = (rFV U(s))VU
and
U((sV)VU) = (rF(VW)V(sV) r
F(VW)W(sW))V,WU.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
15/236
1.1. Abelian presheaves and sheaves 3
A morphism of abelian sheaves is a morphism of the underlying abelian
presheaves. With this notion of morphisms, abelian sheaves form a full
additive subcategory of Psh(X). We denote it by Shv(X).Examples 1.1.4.
(a) Since continuity is a local property, the abelian presheafC0,X is clearlyan abelian sheaf.
(b) The abelian presheaf Lp,X associated to a Borelian measure on
X is not in general an abelian sheaf. As a matter of fact, the condition
U |f|pdV < +may be satisfied locally on U without holding globally. Note however
that the presheaf Lp,X of functions which are locally in Lp,Xis an abelian sheaf.
(c) IfX is a differential manifold, the presheaves Cp,X and DbpX of smooth
and distributional p-forms are clearly abelian sheaves.
(d) Similarly, on a real analytic manifold X, we have the abelian sheaves
ApX and BpX of analytic and hyperfunction p-forms.
(e) On a complex analytic manifold X, we have the abelian sheaves OX ,OX of holomorphic and antiholomorphic functions and the abeliansheaves pX and
p
Xof holomorphic and antiholomorphicp-forms. We
have also the abelian sheaves C(p,q),X and Db(p,q)X of smooth and distri-
butional (p, q)-forms and the sheaves A(p,q)X and B
(p,q)X of analytic and
hyperfunction (p, q)-forms.
Definition 1.1.5. The stalk at x X of an abelian presheaf F is theabelian group
Fx = limUx
U open
F(U)
where the inductive limit is taken over the set of open neighborhoods of x
ordered by . We denote
rFxU : F(U) Fx
the canonical morphism and often use the shorthand notation sx for
rFxU(s)
when there is no risk of confusion.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
16/236
4 1. Survey of sheaf theory
Remark 1.1.6. Let x X and let F be an abelian presheaf on X. To dealwith elements of Fx, we only have to know that:
(a) for any Fx there is an open neighborhood U of x in X ands F(U) such that = sx;
(b) if U, U are two open neighborhoods of x in X and s F(U), s F(U) then sx = s
x if and only if there is an open neighborhood U
of x such that U U U and s|U = s|U.
Proposition 1.1.7. Let F be an abelian presheaf on X. Define A(F)(U)to be the subgroup ofxUFx formed by elements which are locally inF (i.e. such that for any x0 U there is a neighborhood U0 of x0 in Uand s F(U0) with sx = x for any x U0). Turn A(F) into an abelian
presheaf by setting r
A(F)V U ()
x
= x
for anyx V and consider the morphism
a : F A(F)
defined by setting
[a(U)(s)]x = sx
for any x U. Then, A(F) is an abelian sheaf and for any abelian sheaf
G and any morphism g : F G there is a unique morphism g making thediagram
FaGG
g 33hhhh
h A(F)g
1
1
G
commutative. Moreover, a induces an isomorphism
ax : Fx A(F)x
for anyx X.
Definition 1.1.8. We call A(F) the abelian sheaf associated to F.
Examples 1.1.9.
(a) The abelian sheafLp,X considered in Examples 1.1.4 is isomorphicto A( Lp,X).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
17/236
1.1. Abelian presheaves and sheaves 5
(b) To any abelian group M, we may associate the constant presheaf
U M.
This presheaf is in general not a sheaf. We denote its associated sheaf
by MX and call it the constant sheaf with fiber M. For any open
subset U of X, we have
MX(U) = { : U M : locally constant}.
Proposition 1.1.10. The category Shv(X) is abelian. The kernel of amorphism
f : F Gis the abelian sheaf
U Ker f(U);
its cokernel is the abelian sheaf associated to the abelian presheaf
U Coker f(U).
Proposition 1.1.11. A sequence of abelian sheaves
0 Ff
Gg
H 0
is exact if and only if the sequence of abelian groups
0 Fxfx Gx
gx Hx 0
is exact for any x X.
Examples 1.1.12.
(a) Exponential sequence. Let C0,X denotes the (multiplicative) abelian
sheaf formed by non vanishing continuous complex valued functions.Denote
exp : C0,X C0,X
the morphism which sends a continuous complex valued function f to
exp f and denote2i : ZX C0,X
the morphism which sends a locally constant integer valued function
n to a complex valued function 2in. Then, it follows from the local
existence on C of the complex logarithm that
0 ZX2i
C0,Xexp
C0,X 0
is an exact sequence of sheaves.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
18/236
6 1. Survey of sheaf theory
(b) de Rham sequences. Let X be a differential manifold of dimension n
and let d denotes the exterior differential. Working by induction on
n, it is relatively easy to show that for any convex open subset U ofRn the sequence
0 CRn (U) C0,Rn (U)
d
d Cn,Rn (U) 0
is exact. This result, often referred to as the Poincare lemma, shows
directly that
0 CX C0,X
d
d Cn,X 0
is an exact sequence of abelian sheaves. Similar results hold with C,Xreplaced by DbX (and by AX or BX if X is a real analytic manifold).
(c) Dolbeault sequences. Let X be a complex analytic manifold. Then,
for any smooth (p, q)-form we have
d = +
with (resp. ) of type (p + 1, q) (resp. (p, q+ 1)). This gives riseto morphisms
: C(p,q),X C
(p+1,q),X , : C
(p,q),X C
(p,q+1),X
such that 2 = 0, 2
= 0, + = 0. If U is a convex open subset
ofCn, one checks by induction on n that the sequences
0 p
Cn (U) C(p,0)
,Cn (U)
C(p,n)
,Cn (U) 0
and
0 pC
n (U) C(0,p),Cn (U)
C
(n,p),Cn (U) 0
are exact. Therefore, we see that
0 pX C(p,0),X
C(p,n),X 0
and0 p
X C
(0,p),X
C
(n,p),X 0
are exact sequences of abelian sheaves.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
19/236
1.2. Sections of an abelian sheaf 7
Proposition 1.1.13.
(a) Let (Ix)xX be a family of injective abelian groups. Then, the abelian
sheaf I defined by setting
I(U) =xU
Ix
for any open subset U of X andrIV U(s)
x
= sx
for any open subset V of U and anyx V is injective.(b) Let F be an abelian sheaf on X. Then, there is a monomorphism
F I
where I is a sheaf of the type considered in (a). In particular, theabelian category Shv(X) has enough injective objects.
Remark 1.1.14. As a consequence of the preceding proposition, we get
that any functor F : Shv(X) A where A is an abelian category has aright derived functor. Note that, in general, Shv(X) does not have enoughprojective objects.
1.2 Sections of an abelian sheaf
Definition 1.2.1. Let A be a subset of X and let F be an abelian sheafon X.
A section of F on A is an element
xA
Fx
with the property that that for any x0 A there is an open neighborhoodU0 of x0 in X and s F(U0) such that
x = sx
for any x A U0. When A = X, we call sections ofF on A global sectionof F.
The support of a section of F on A is the set
supp() = {x A : x = 0}.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
20/236
8 1. Survey of sheaf theory
It is the relatively closed subset of A.
Sections of F on A form an abelian group that we denote by (A; F).
Let B be a subset of X such that B A and let (A; F). Then,rFBA() is the element of (B; F) defined by settingrFBA()
x
= x
for any x B. We will often use |B as a shorthand notation for rFBA().
Of course, we have
supp(|B) = supp() B.
Remark 1.2.2. Let U be an open subset of X and let F be an abeliansheaf on X. Then, the canonical morphism
F(U) (U; F)
which sends s F(U) to (sx)xU is an isomorphism. Hereafter, we willoften use this isomorphism to identify F(U) and (U; F) without furthernotice. Note that if V is an open subset of U, the two definitions of rFV Uare compatible with this identification.
Definition 1.2.3. Let X be a topological space and let A be a subspace ofX.
We say that A is relatively Haussdorf in X if for any x = y in A we can
find open neighborhoods U and V of x and y in X such that U V = .
By an open covering of A in X, we mean a set U of open subsets of Xsuch that for any x A there is U U containing x. Such a covering islocally finite on A if any x A has a neighborhood V in X for which the
set
{U U : U V = }is finite.
We say that A is relatively paracompact in X if it is relatively Haussdorf
and if for any open covering U of A in X we can find an open covering Vof A in X which is locally finite on A and such that for any V V there isU U with V U.
Remark 1.2.4. One checks easily that a subspace A of a topological space
X is relatively paracompact if it has a fundamental system of paracompact
neighborhoods. This will be the case in particular in the following cases:
(a) X is completely paracompact (e.g. metrizable);
(b) A is closed and X is paracompact.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
21/236
1.3. Cohomology with supports 9
Other examples of relatively paracompact subspaces are given by relatively
Haussdorf compact subspaces.
Proposition 1.2.5. Let A be a relatively paracompact subspace of X andlet F be an abelian sheaf on X. Then,
(a) for any (A; F) there is an open neighborhoodU of A in X ands (U; F) such that s|A = ;
(b) ifU, U are two open neighborhoods of A in X and s (U; F), s (U, F) then s|A = s
|A if and only if there is an open neighborhood
U of A such that U U U and s|U = s|U .
In other words, we have
(A; F) = limUAU open
(U; F).
Proof. See [24].
Proposition 1.2.6. Let A be a topological subspace of X and let F be anabelian sheaf on X. Then, the abelian presheaf F|A defined by setting
F|A(U) = (U; F)
for any open subset U of A and
rF|AV U = r
FVU
for any chain V U of open subsets of A is an abelian sheaf.
Remark 1.2.7. It follows from the preceding results that sections of F onA may be considered as the global sections of the abelian sheaf F|A.
1.3 Cohomology with supports
Definition 1.3.1. A family of supports of X is a set of closed subsets of
X such that
(a) ifF1 is a closed subset of X included in F2 , then F1 ;
(b) for any F1, F2 , there is F3 such that F1 F2 F3.
Let F be an abelian sheaf on X. Then global sections s of F such that
supp(s)
form an abelian group that we denote (X; F).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
22/236
10 1. Survey of sheaf theory
Examples 1.3.2.
(a) The set X of all the closed subsets ofX is clearly a family of supports
and we haveX (X; F) = (X; F)
for any abelian sheaf F.
(b) Let F be a closed subset of X. Then, the set F of all the closed
subsets of F is a family of supports. In this case we set for short
F(X; F) = F (X; F)
for any abelian sheafF. Note that this special case allows us to recoverthe general one. As a matter of fact, we have
(X; F) = limF
F(X; F).
(b) Let X be a Haussdorf space. Then, the set c of all compact subsets
of X is a family of supports. We set for short
c
(X; F) = c
(X; F)
for any abelian sheaf F.
Let be a family of supports of X.
Proposition 1.3.3. The functor
(X; ) : Shv(X) Ab
is left exact and has a right derived functor
R(X; ) : D+(Shv(X)) D+(Ab).
Proof. The left exactness follows directly from the structure of kernels
in Shv(X). The existence of the right derived functor follows from Re-mark 1.1.14.
Definition 1.3.4. Let F be an abelian sheaf on X. We define the k-thcohomology group of X with coefficients in F and supports in as the
group Hk[R(X; F)].
To lighten notations, we denote it by
Hk(X; F).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
23/236
1.4. Flabby and soft abelian sheaves 11
If is the family of all closed subsets of X, we shorten the notation by
dropping the . Similarly, if is the family F (resp. c) considered in
Examples 1.3.2, we replace it by F (resp. c).An abelian sheaf F is -acyclic if Hk(X; F) = 0 for any k > 0.
Remark 1.3.5. Let F be an abelian sheaf on X. By a well-known resultof homological algebra
R(X; F) (X; R)
if R is a right resolution of F by -acyclic sheaves. The aim of the nextsection is to give basic examples of such sheaves
1.4 Flabby and soft abelian sheaves
Definition 1.4.1. An abelian sheaf F on X is flabby if for any chain U1 U2 of open subsets of X
rFU1U2 : (U2; F) (U1; F)
is an epimorphism.
Examples 1.4.2.
(a) Any injective abelian sheaf is flabby.
(b) Let (Mx)xX be a family of abelian groups. Then,
U xU
Mx
is a flabby sheaf.
(c) The sheaf Lp,X of Examples 1.1.4 is flabby.
(d) The sheaf BX of hyperfunctions is flabby.
Proposition 1.4.3. A flabby abelian sheaf F is -acyclic for any familyof supports . Moreover, if in an exact sequence of abelian sheaves
0 F G H 0
F and G are flabby, then so is H.
Definition 1.4.4. An abelian sheaf F on X is -soft if for any chainF1 F2 of
rFF1F2 : (F2; F) (F1; F)
is an epimorphism.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
24/236
12 1. Survey of sheaf theory
Examples 1.4.5.
(a) Assume X is normal. Thanks to Urysohns extension result, it is clear
that the abelian sheaf C0,X is -soft for any family of supports .
(b) IfX is a differential manifold, it follows from the existence of partitions
of unity that the abelian sheaves Cp,X and DbpX are -soft for any
family of supports . The same is true of the sheaves C(p,q),X and Db(p,q)X
if X is a complex analytic manifold.
Definition 1.4.6. A family of supports is paracompactifying if
(a) any F is paracompact;
(b) for any F1 , there is F2 with F1
F2.
Examples 1.4.7.
(a) If X is paracompact, then the family formed by the closed subsets of
X is paracompactifying.
(b) IfX is a locally compact space, then the family formed by the compact
subsets of X is paracompactifying.
Proposition 1.4.8. Assume the abelian sheaf F is -soft and the family is paracompactifying. Then, F is -acyclic. Moreover, if in the exactsequence of abelian sheaves
0 F G H 0
F and G are -soft, then so is H.
Corollary 1.4.9 (de Rham and Dolbeault theorems).
(a) For any differential manifoldX, we have the canonical isomorphisms
Hk(X;CX) Hk((X; C,X)) H
k((X; DbX))
and
Hkc (X;CX) Hk(c(X; C
,X)) H
k(c(X; DbX))
for anyk N.
(b) For any complex analytic manifoldX, we have the canonical isomor-
phisms
Hk(X; pX) Hk((X; C
(p,),X)) Hk((X; Db
(p,)X ))
and
Hkc (X; pX) H
k(c(X; C(p,),X)) H
k(c(X; Db(p,)X ))
for anyk, p N.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
25/236
1.5. Cohomology of subspaces and tautness 13
Proof. Thanks to Examples 1.4.5 (b) and Examples 1.4.7, this follows di-
rectly from Remark 1.3.5 and Proposition 1.4.8.
Exercise 1.4.10. Let Bn denote the open unit ball ofRn. Show that for
n 1
Hk(Bn;CBn )
C if k = 0
0 otherwise
Solution. Recall that for any convex open subset U of Rn, the Poincare
lemma for d shows that the sequence
0 (U;CRn
) (U; C0
,Rn) (U; Cn
,Rn ) 0
is exact. Therefore, by the de Rham theorem we have
Hk(U;CU)
C if k = 0
0 otherwise
Since Bn is convex, the conclusion follows.
1.5 Cohomology of subspaces and tautness
Definition 1.5.1. Let X be a topological space and let be a family of
supports on X. We say that a subspace A of X is -taut if the canonical
morphism
limUAU open
HkU(U; F|U) HkA(A; F|A)
is an isomorphism for any k 0 and any abelian sheaf F on X.
Remark 1.5.2. It is easily seen that a subspace A of X is -taut if and
only if for any flabby sheaf F on X
(a) the abelian sheaf F|A is A-acyclic;
(b) the canonical morphism
(X; F) A(A; F|A)
is surjective.
In this case, it follows that
F R(A; F|A)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
26/236
14 1. Survey of sheaf theory
is the right derived functor of
F (A; F).
One should however be aware that this result is false in general.
Proposition 1.5.3. Let X be a topological space and let be a family of
supports on X. AssumeA is a topological subspace ofX. Then A is-taut
in the following cases:
(a) is arbitrary and A is open;
(b) is paracompactifying and A is closed;
(c) is paracompactifying and X is completely paracompact;
(d) is the family of all closed subset X and A is both compact and
relatively Haussdorf.
Exercise 1.5.4. Let Bn denote the closed unit ball ofRn. Show that for
n 1
Hk(Bn;CBn ) C if k = 0
0 otherwise
Solution. Since Bn has a fundamental system of neighborhoods formed by
open balls of Rn, this follows directly from Proposition 1.5.3 and Exer-
cise 1.4.10
1.6 Excision and Mayer-Vietoris sequences
Definition 1.6.1. Let A be a subset ofX and let be a family of supports
of X. We set|A = {F : F A}
and
A = {F A : F }.
Remark 1.6.2. Let A be a subset of X and let a family of supports of
X. Clearly, for any abelian sheaf F on X, we have canonical morphisms
|A (X; F) (X; F)
and
(X; F) A(A; F)
which induce similar morphisms at the level of derived functors.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
27/236
1.6. Excision and Mayer-Vietoris sequences 15
Proposition 1.6.3 (Excision). Let A be a subset of X and let be a
family of supports of X. Assume either A is open or A is closed and is
paracompactifying. Then, for any object F
of D
+
(Shv(X)), we have thecanonical distinguished triangle
R|X\A (X; F) R(X; F
) RA(A; F)
+1 .
In particular, for any abelian sheaf F, we have the excision long exactsequence:
Hk|X\A (X; F)GG Hk(X; F)
GG HkA(A; F) -,*+
/.()GG Hk+1|X\A (X; F)
GG Hk+1 (X; F)GG Hk+1A(A; F)
Proof. Let us recall the proof of this result since a detailed understanding
of its mechanism will be necessary in various parts of this book. We treat
only the case where A is open; the other case being similar.
Assume G is a flabby sheaf and let s (A; G) be such that
supp(s) F A
with F in . The zero section on X\ F and the section s on A coincide on(X \ F) A = A \ F. Therefore, there is a section s of G on (X \ F) Asuch that s|X\F = 0, s
|A = s. Since G is flabby and (X\ F) A is open, we
may extend s into a section s of G on X. For this section, we have
s|X\F = 0, s|A = s.
Hence, supp s F and belongs to . These considerations show that
(X; G) A(A; G)
is an epimorphism. Moreover, a simple computation shows that
0 |X\A (X; F) (X; F) A(A; F)
is exact for any abelian sheaf F. It follows that if G is a flabby resolutionof the complex F, then
0 |X\A (X; G) (X; G) A(A; G) 0
is an exact sequence of complexes of abelian groups. Since flabby sheaves are
acyclic for the various functors involved, we get the announced distinguished
triangle. The last part of the result follows from the snakes lemma.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
28/236
16 1. Survey of sheaf theory
Remark 1.6.4. Let us recall that the snakes lemma states that an exact
sequence of complexes of abelian groups
0 Au
Bv
C 0
induces a long exact sequence of cohomology
Hk(A)Hk(u)
GG Hk(B)Hk(v)
GG Hk(C) -,*+/. k()
GG Hk+1(A)Hk+1(u)
GG Hk+1(B)Hk+1(v)
GG Hk+1(C)
where k is defined as follows. Let [ck] be a cohomology class in Hk(C).
Since vk is surjective, there is bk Bk such that
vk(bk) = ck.
Using the fact that dk(ck) = 0, one sees that dk(bk) is in Ker vk+1. Hence,
there is ak+1 Ak+1 such that uk+1(ak+1) = dk(bk). The cohomology classof ak+1 is the image of [ck] by k. A way to remember this definition is to
follow the dotted path in the following diagram:
Akuk GG
Bkvk GG
00
1
W
dt
~Ck
Ak+1uk+1 GG
dt
Bk+1
vk+1 GG Ck+1
Proposition 1.6.5 (Mayer-Vietoris). Let A, B be two subsets ofX andlet be a family of supports on X. Assume that either A and B are open
orA andB are closed and is paracompactifying. Then, for any object F
of D+(Shv(X)), we have the canonical distinguished triangle
R(AB)(A B; F) RA(A; F
) RB(B; F) R(AB)(A B; F
)+1
In particular, ifF is an abelian sheaf, we have the Mayer-Vietoris long exactsequence
Hk(AB)(A B; F)GG HkA(A; F) H
kB(B; F)
GG Hk(AB)(A B; F) -,*+/.()
GG Hk+1(AB)(A B; F)GG Hk+1A(A; F) H
k+1B(B; F)
GG Hk+1(AB)(A B; F)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
29/236
1.6. Excision and Mayer-Vietoris sequences 17
Proof. The proof being similar to that of Proposition 1.6.3, we will not recall
it in details. We only recall that it is based on the fact that the sequence
0 (AB)(A B; G)
A(A; G) B(B; G)
(AB)(A B; G) 0
where (s) = (s|A, s|A), (s, s) = s|AB s
|AB is exact when A, B are
open and G is flabby or when A, B are closed, is paracompactifying and
G is -soft.
Exercise 1.6.6.
(a) Let Sn denotes the unit sphere in Rn+1. By using Mayer-Vietoris
sequence and de Rham theorem, show that for n 1
Hk(Sn ;CSn )
C if k = 0, n
0 otherwise
(b) Show by excision that for n 1
Hk
c(Bn;CB
n
) C if k = n0 otherwiseSolution. (a) Assume n 1. Set
S+n = {x Sn : xn+1 0}, Sn = {x Sn : xn+1 0}
and identify Sn1 with S+n S
n (see figure 1.6.1). Since Sn = S
+n S
n , we
have the Mayer-Vietoris long exact sequence
Hk(Sn;C) GG Hk(S+n ;C) H
k(Sn ;C)GG Hk(Sn1;C) -,*+
/.()GG Hk+1(Sn;C) GG H
k+1(S+n ;C) Hk+1(Sn ;C)
GG Hk+1(Sn1;C)
Recall that, for any ]0, 1[, {x Sn : xn+1 > } is an open subset of Snwhich is diffeomorphic to a ball ofRn. Therefore, working as in (a), we see
that
Hk
(S+
n ;C) C if k = 00 otherwiseOf course, there is a similar result for Sn . It follows that
Hk+1(Sn;C) Hk(Sn1;C) (*)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
30/236
18 1. Survey of sheaf theory
Figure 1.6.1:
Sn
S+n
Sn1
xn+1
0
if k > 0 and that H0(Sn;C) and H1(Sn;C) are isomorphic to the kernel and
the cokernel of the morphism
H0(S+n ;C) H0(Sn ;C) H
0(Sn1;C) (**)
(, ) |Sn1 |Sn1
Note that since S+n and Sn are clearly connected spaces, the locally constant
complex valued functions and are in fact constant.
Let us assume first that n = 1. Since S0 = {0, 1}, we have
Hk(S0
;C) C2 if k = 0
0 otherwise.
Therefore, the morphism (**) becomes up to isomorphisms the additive
map
C2 C2
(x, y) (x y, x y)
It follows that H1(S1;C) C and that H0(S1;C) C. This last isomor-
phism reflecting the fact that the circle S1 is connected.Assume now that n > 1 and that
Hk(Sn1;C)
C if k = 0, n 1
0 otherwise
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
31/236
1.6. Excision and Mayer-Vietoris sequences 19
The morphism (**) now becomes equivalent to the additive map
C2 C
(x, y) x y
Hence, H1(Sn;C) 0 and H0(Sn;C) C. Moreover, thanks to (*) we see
that for k > 1 we have
Hk(Sn;C)
C if k = n
0 otherwise
The conclusion follows by induction.
(b) Assume n 1. Since Bn \ Bn = Sn1 and since any closed subsetof Bn is compact, we have the excision distinguished triangle
Rc(Bn;C) R(Bn;C) R(Sn1;C)+1
From the associated long exact sequence, we deduce that
Hk(Sn1;C) Hk+1c (Bn;C)
for k > 0 since in this case
Hk(Bn;C) Hk+1(Bn;C) 0.
Moreover, we see also that H0c(Bn;C) and H1c(Bn;C) are the kernel and
cokernel of the morphism
H0(Bn;C) H0(Sn1;C)
|Sn1
For n = 1, this morphism is equivalent to
C C2
x (x, x)
and we get H0c(B1;C) 0 and H1c(B1;C) C.
For n > 1, it becomes equivalent to
C Cx x
and we get H0c(Bn;C) H1c(Bn;C) 0. The conclusion follows by induc-
tion on n.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
32/236
20 1. Survey of sheaf theory
Exercise 1.6.7. Let I = [a, b] be a compact interval ofR. Show by usingtautness and a suitable Mayer-Vietoris sequence that for any abelian sheaf
F on I, we have Hk(I; F) = 0
for k > 1. Establish also that this relation holds for k = 1 if
(I; F) Fx
is an epimorphism for every x I. As an application compute
H(I; MI)
for any abelian group M.
Solution. Fix k > 0 and assume there is c Hk(I; F) which is non-zero.Set
x0 = inf{x I : c|[a,x] = 0}.
Since by tautness
limx>x0
Hk([a, x]; F) Hk([a, x0]; F)
we see that c|[a,x0] = 0 and hence that x0 > a. Since
limx 1 or k = 1 and
(I; F) Fx
is an epimorphism, it follows from this sequence that
Hk([a, x0]; F)
Hk([a, x]; F) Hk([x, x0]; F).
This gives us a contradiction since both c|[a,x] and c|[x,x0] are 0 although
c|[a,x0] = 0.
The application to F = MI is obvious. We get
Hk(I; MI) =
M if k = 0;
0 otherwise.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
33/236
1.7. Inverse and direct images 21
1.7 Inverse and direct images
Definition 1.7.1. Let f : X Y be a morphism of topological spaces and
let G be an abelian sheaf on Y.Let U be an open subset of X. We define f1(G)(U) as the abelian
subgroup of
xUGf(x) formed by the elements such that for any x0 U
there is an open neighborhood U0 of x0 in U, an open neighborhood V0 of
f(U0) and s G(V0) such that x = sf(x) for any x U0. Clearly,
U f1(G)(U)
is an abelian sheaf on X. We call it the inverse image of G by f. Note that
by construction there is a canonical pull-back morphism
f : (Y; G) (X; f1(G)).
Remark 1.7.2. It follows at once from the preceding definition that we
may identify f1(G)x and Gf(x). In particular, the functor
f1 : Shv(Y) Shv(X)
is exact.
Proposition 1.7.3. Let f : X Y and g : Y Z be two morphisms oftopological spaces. Then,
f1(g1(H)) (g f)1(H)
canonically and functorially for H Shv(Z). Moreover,
(id1X )(F) F
canonically and functorially for F Shv(X).
Examples 1.7.4.
(a) Let aX : X {pt} be the morphism which maps the topological spaceX to a point. Identifying sheaves on {pt} with their global sections,we have
MX a1X (M)
for any abelian group M.
(b) Combining the preceding proposition with example (a), we get a
canonical isomorphismf1(MY) MX
for any abelian group M and any morphism of topological spaces
f : X Y.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
34/236
22 1. Survey of sheaf theory
(c) Let i : A X be the canonical inclusion of a subspace A of X andlet F be an abelian sheaf on X. One checks easily that
i1(F) = F|A.
Definition 1.7.5. We say that (X; F) is an abelian sheafed space if X is atopological space and F is an abelian sheaf on X. We say that
(f; ) : (X; F) (Y; G)
is a morphism of abelian sheafed spaces if (X; F) and (Y; G) are abeliansheafed spaces, f : X Y is a morphism of topological spaces and
: f1G F
is a morphism of sheaves.
Examples 1.7.6.
(a) Let f : X Y be a morphism of topological spaces. Then thecanonical isomorphism
f : f1MY MX
gives rise to a morphism of sheafed spaces
(f; f) : (X; MX) (Y; MY).
(b) Let f : X Y be a morphism of topological spaces. Then,
g g f
induces a canonical morphism of abelian sheaves
f : f1C0,Y C0,X
and hence a morphism of sheafed spaces
(f; f) : (X; C0,X) (Y; C0,Y).
(c) Similarly, a morphism of differential manifolds f : X Y inducesmorphisms of abelian sheaves
f : f1Cp,Y Cp,X
corresponding to the pull-back of differential p-forms. Hence, we have
a canonical morphism of sheafed spaces
(f; f) : (X; Cp,X) (Y; Cp,Y).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
35/236
1.7. Inverse and direct images 23
Proposition 1.7.7. A morphism of sheafed spaces
(f; ) : (X; F) (Y; G)
induces a morphism
R(Y; G) R(X; F)
and, in particular, a canonical morphism
(f; ) : H(Y; G) H(X; F).
Moreover, these morphisms are compatible with the composition of mor-
phisms of sheafed spaces.
Proof. We have a functorial morphism
(Y; G) (X; f1G) (*)
Since the functor f1 is exact, a standard result of homological algebra
gives us a morphism
R(Y; G) R(X; f1G).
Composing with the canonical morphism
R(X; f1G) R(X; F)
induced by , we get the requested morphism. It is possible to visualize
this abstract construction more explicitly as follows. Assume we are given
a quasi-isomorphism
G
J
where J is a complex of (Y; )-acyclic sheaves and a commutative diagramof the form
f1(G) f1 ()GG
f1(J)
F
GG I
where is a quasi-isomorphism and I is a complex of (X; )-acyclicsheaves. Then, in D+(Ab), we have the commutative diagram
R(Y; G) /o(1)
GG
(2)
(Y; J)
(3)
(X; f1(J))
(4)
R(X; F) /o(5)
GG (X;I)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
36/236
24 1. Survey of sheaf theory
where (1) and (5) are induced by and , (3) is induced by morphism of
the type (*), (4) is induced by and (2) is the morphism defined abstractly
above.Examples 1.7.8.
(a) Let f : X Y be a morphism of topological spaces. By applying thepreceding proposition to the canonical morphism of sheafed spaces
(f; f ) : (X; MX) (Y; MY)
we get a canonical morphism
R(Y; MY) R(X; MX)
in D+(Ab). The associated morphism
(f; f) : H(Y; MY) H(X; MX)
is often simply denoted by f. Clearly, f = id if f is the identity of
X. Moreover, ifg : Y Z is another morphism of topological spaces,we have f g = (g f). This shows that X H(X; MX) is a
functor on the category of topological spaces. In particular, X Y
implies H(X; MX) H(Y; MY).
(b) Let f : X Y be a morphism of differential manifolds. It is wellknown that the pull-back of differential forms is compatible with the
exterior differential. In other words, we have
d(f()) = f(d)
for any (Y; Cp,Y). Using what has been recalled in the proof
of the preceding proposition, we see that we have the commutativediagram
Hp(Y;CY)f
GG
O
Hp(X;CX)O
Hp((Y; C,Y)) fGG Hp((X; C,X))
where the first horizontal arrow is the one defined in (a) and the
two vertical isomorphisms come from the de Rham theorem. Thanks
to what has been said in (a), we see also that, up to isomorphism,
de Rham cohomology of X
H((X; C,X))
depends only on the topology ofX and not on its differential structure.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
37/236
1.7. Inverse and direct images 25
Definition 1.7.9. Let f : X Y be a morphism of topological spaces andlet F be an abelian sheaf on X.
The direct image of the abelian sheaf F by f is the sheaf f(F) on Ydefined by settingf(F)(V) = F(f1(V))
for any open subset V of Y; the restriction morphisms being the obvious
ones.
Example 1.7.10. Let aX : X {pt} be the morphism which maps thetopological space X to a point. Identifying sheaves on {pt} with theirabelian group of global sections, we have
aX(F) (X; F)
for any abelian sheaf F on X.
Proposition 1.7.11 (Adjunction formula). We have a canonical func-
torial isomorphism
Hom Shv(Y)(G, f(F)) Hom Shv(X)(f1(G), F).
This isomorphism is induced by two canonical functorial morphisms
f1(f(F)) F
and
G f(f1(G)).
Proof. We will only recall the construction of the canonical functorial mor-
phism
Hom Shv(Y)(G, f(F)) Hom Shv(X)(f1(G), F) (*)
Let h : G f(F) be a morphism of abelian sheaves. For any open subset
V of Y, we get a morphism
h(V) : G(V) F(f1(V))
and using Remark 1.1.6, it is easy to deduce from these morphisms a mor-
phism
hx,f(x) : Gf(x) Fx
for any x X. Now, let U be an open subset of X and let f1(G)(U).One checks directly that (hx,f(x)(x))xU F(U). Hence, for any open
subset U of X, we have a morphism
h(U) : f1(G)(U) F(U).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
38/236
26 1. Survey of sheaf theory
These morphisms give rise to the morphism
h : f1(G) F
image of h by (*).
Proposition 1.7.12. Let f : X Y and g : Y Z be two morphisms of
topological spaces. Then,
g(f(F)) (g f)(F)
canonically and functorially for F Shv(X). Moreover,
(idX)(F) F
canonically and functorially for F Shv(X).
Proposition 1.7.13. The direct image functor
f : Shv(X) Shv(Y)
is left exact and has a right derived functor
Rf : D+(Shv(X)) D+(Shv(Y)).
Moreover, if g : Y Z is another morphism of topological spaces, then
Rg Rf R(g f).
Example 1.7.14. Denoting aX : X {pt} the canonical map, we deducefrom the functorial isomorphism
aX(F) (X; F)that
RaX() R(X; ).
Therefore, the second part of the preceding proposition contains the fact
that
R(Y; Rf(F)) R(X; F);
a result which replaces Lerays spectral sequence in the framework of derived
categories.Proposition 1.7.15. Assume that
(a) the map f is closed (i.e. such that f(F) is a closed subset of Y if F is
a closed subset of X);
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
39/236
1.7. Inverse and direct images 27
(b) the fiber f1(y) is a taut subspace of X for anyy f(Y).
Then,
[Rf(F)]y = R(f1(y); F)for anyy Y and any abelian sheaf F on X.
Example 1.7.16. Thanks to Proposition 1.5.3, it is clear that the condi-
tions on f in the preceding proposition are satisfied if one of the following
conditions holds :
(a) f is closed and X is metrizable;
(b) f is closed, Y is Haussdorf and X is paracompact;
(c) f is proper and X is Haussdorf.
Remark 1.7.17. Note that under the assumptions of Proposition 1.7.15,
we have of course
[f(F)]y = (f1(y); F)
but that this formula may be false in general.
Corollary 1.7.18 (Vietoris-Begle). Assume that
(a) the map f is closed and surjective,
(b) the fiber f1(y) is a taut subspace of X for anyy Y,
(c) the fiber f1(y) is connected and acyclic (i.e.
Hk(f1(y); Mf1 (y)) 0
for anyk > 0 and any abelian group M) for any y Y.
Then, the canonical morphism
G Rf(f1
(G))
is an isomorphism for any G D+(Shv(Y)). In particular, the canonicalmorphism
R(Y; G) R(X; f1G)
is an isomorphism in D+(Ab).
Proof. It is sufficient to note that
(f1G)|f1 (y) (Gy)f1 (y)
and that the canonical morphism
M H0(f1(y); Mf1 (y))
is an isomorphism if f1(y) is non-empty and connected.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
40/236
28 1. Survey of sheaf theory
Figure 1.7.1:
x0
(x1, , xn)
x(U X)
x(U)
Remark 1.7.19. Let us recall a few facts about Stokes theorem which
are needed in the following exercise. Let X be an oriented n-dimensional
differential manifold with boundary. As is well-known, the orientation ofX
induces an orientation on X. This orientation is characterized by the fact
that if x : U Rn is a positively oriented local coordinate system of X onan open neighborhood of u X such that
x(u) = 0, x(U) = {x Bn : x0 0}, x(U X) = {x Bn : x0 = 0};
(see Figure 1.7.1) then (x1, , xn)|UX is a positively oriented coordinatesystem of X. With this orientation ofX, Stokes formula states that
X
d =
X
for any c(X; Cn1,X ).
Exercise 1.7.20.
(a) Let X be an oriented n-dimensional differential manifold. Show that
integration of smooth compactly supported n-forms induces a mor-
phism X
: Hnc (X;C) C
(b) Let X be an orientedn-dimensional differential manifold with bound-
ary. Express the various morphisms of the excision long exact sequence
Hkc (X \ X;C)uk GG Hkc (X;C)
vk GG Hkc (X;C) -,*+/. k()
GG Hk+1c (X\ X;C)uk+1 GG Hk+1c (X;C)
vk+1 GG Hk+1c (X;C)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
41/236
1.7. Inverse and direct images 29
in terms of de Rham cohomology. Show in particular that
X n1cn1 = X cn1for anycn1 Hn1c (X;C).
Solution. (a) Integration gives us a morphismX
: c(X; Cn,X) C.
By Stokes theorem, we know thatX
d =X
= 0
for any c(X; Cn1,X ). Therefore,
X
induces a morphismX
: c(X; Cn,X)/dc(X; C
n,X) C
and the conclusion follows from the de Rham theorem.
(b) Since C,X is a c-soft resolution of the sheaf CX , the long exactsequence of cohomology comes from the application of the snakes lemma
to the exact sequence of complexes
0 c(X \ X; C,X |X\X
) c(X; C,X ) c(X; C
,X |X
) 0
The canonical restriction morphism
C,X |X\X C,X\X
is clearly an isomorphism. As for the restriction morphism
C,X |X C,X
it is a quasi-isomorphism since in the commutative diagram of complexes
CX |X/o GG
CX
C
,X |X GG C
,X
both vertical arrows are quasi-isomorphisms. It follows that
c(X \ X; C,X |X\X
) c(X \ X; C,X\X )
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
42/236
30 1. Survey of sheaf theory
and that
c(X; C,X |X
) qis
(X; C,X).
Computation of uk. Let ck Hkc (X \ X;C) be represented by
k c(X \ X; Ck,X\X ).
By extension by zero, k becomes a class
k
c(X; Ck,X )
and uk(ck) is represented by k.
Computation of v
k
. Let c
k
H
k
c (X;C
) be represented byk c(X; C
k,X)
and let j : X X denotes the inclusion map. Then, vk(k) is representedby
j(k) c(X; Ck,X).
Computation of k. Let ck Hk(X;C) be represented by
k c(X; Ck,X).
Set K = supp k. We know from elementary differential geometry (collar
neighborhood theorem) that there is a neighborhood U of X in X, a
differentiable map p : U X such that p j = idX and a smoothfunction equal to 1 on a neighborhood of X and such that supp is
a p-proper subset U. Denote k
the k-form on X obtained by extending
p by zero outside of U. Clearly, k
has compact support and
jk
= k.
Therefore, it follows from the snakes lemma that k(ck) is represented by
dk|X\X. As expected, this form has compact support. As a matter of
fact,
dk|U = d p
k + pdk = d pk
and d = 0 in a neighborhood of X. Assuming now that k = n 1, wesee using Stokes theorem that
X\X n1cn1 = X dn1 = X n1 = X jpn1=
X
n1 =
X
cn1.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
43/236
1.8. Homotopy theorem 31
1.8 Homotopy theorem
Definition 1.8.1. Let F, G be abelian sheaves on X and Y respectively
and let
(h; ) : (X I;p1X F) (Y; G)
be a morphism where I = [0, 1] and pX : X I X denotes the firstprojection. Let t [0, 1]. We denote
it : X X I
the morphism defined by setting it(x) = (x, t) and ht the morphism h it.Applying i1t to
: h1G p1X F
we get a morphism
i1t : i1t h
1G i1t p1X F.
Since h it = ht and pX it = idX , this gives us a morphism
t : h1t G F
and a corresponding morphism
(ht; t) : (X; F) (Y; G).
We call (h; ) a homotopy between
(h0; 0) : (X; F) (Y, G)
and
(h1; 1) : (X; F) (Y; G).
Two morphisms of abelian sheafed spaces connected by a homotopy are said
to be homotopic.Proposition 1.8.2. Let F, G be abelian sheaves on X andY. Assume themorphisms
(f0; 0) : (X; F) (Y; G)
and
(f1; 1) : (X; F) (Y; G)
are homotopic. Then, the morphisms
(f0; 0)
: H
(Y; G) H
(X; F)
and
(f1; 1) : H(Y; G) H(X; F)
are equal.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
44/236
32 1. Survey of sheaf theory
Proof. Since the application
pX : X I X
is proper and surjective and I is connected and acyclic (see Exercise 1.6.7),
Corollary 1.7.18 shows that the canonical morphism
(pX ; X) : (X I;p1X F) (X; F)
induces the isomorphism
(pX ; X) : H(X; F)
H(X I;p1X F).
If (it; t) : (X; F) (X I;p1X F)
denotes the canonical morphism, we have
(pX ; X) (it; t) = id .
It follows that (it; t) is the inverse of the isomorphism (pX ; X)
and thus
does not depend on t [0, 1].Let
(h; ) : (X I;p
1
X F) (Y; G)be a homotopy between (f0; 0) and (f1; 1). Since
(h; ) (i0; 0) = (f0; 0), (h; ) (i1; 1) = (f1; 1),
we see that
(f0 ; 0) = (i0; 0)
(h; ) = (i1; 1) (h; ) = (f1; 1)
.
Corollary 1.8.3. If two morphisms
f0 : X Y and f1 : X Y
are homotopic in the topological sense, then, for any abelian group M,
f0 : H(Y; MY) H
(X; MX)
and
f1 : H(Y; MY) H
(X; MX)
are equal. In particular, if X and Y are homotopically equivalent, then
H(X; MX) H(Y; MY)
for any abelian group M.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
45/236
1.8. Homotopy theorem 33
Examples 1.8.4.
(a) The preceding corollary allows us to show that
Hk(Bn; MBn ) =M if k = 0
0 otherwise
for any abelian group M. As a matter of fact,
h : Bn I Bn
(x, t) tx
is a homotopy between the constant map
h0 : Bn Bn
x 0
and the identity map
h1 : Bn Bn
x x
Therefore, the inclusion map j0
: {0} Bn
and the projection map
p0 : Bn {0} are inverse of each other in the category of topologicalspaces modulo homotopy. Hence,
H(Bn; MBn ) H({0}; M{0})
and the conclusion follows.
Note that in contrast with Exercise 1.4.10, we have not made use of
de Rham theorem. A similar reasoning shows that
Hk(Bn; MBn ) M if k = 0
0 otherwise
for any abelian group M.
(b) Working as in Exercise 1.6.6 (a) and (b), we can deduce from (a) that,
for n 1 and any abelian group M, we have
Hk(Sn; MSn ) M if k = 0, n
0 otherwiseand
Hkc (Bn; MBn )
M if k = n
0 otherwise
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
46/236
34 1. Survey of sheaf theory
Exercise 1.8.5. Assume n 1. EndowBn+1 with the orientation inducedby the canonical orientation ofRn+1 and orient Sn asBn+1. Denote
the
morphism obtained by composing the morphismHn(Sn;Z) H
n(Sn;C)
induced by the inclusion ofZ in C with the integralSn
: Hn(Sn;C) C
of Exercise 1.7.20. Show that the group
Hn(Sn;Z)
has a unique generator vSn such that
vSn = 1 and that
induces an
isomorphism between Hn(Sn;Z) andZ.
Solution. Since Hn(Sn;Z) Z, this group has only two generators g1 andg2 = g1. The uniqueness is thus obvious. Let us prove the existence. Weshall use the same notations as in Exercise 1.6.6. It is clear that we have
the following canonical morphism of distinguished triangles
Rc(Sn;Z) GG Rc(S+n ;Z) Rc(S
n ;Z)
GG Rc(Sn1;Z)+1
GG
Rc(S+n \ Sn1;Z)
GG
yy
Rc(S+n ;Z)
GG
( id0 )yy
Rc(Sn1;Z)+1
GG
idyy
where the first vertical arrow is induced by zero extension of sections. It
follows that we have the commutative diagram
H
n1
(Sn1;Z)
n1 GG
H
n
(Sn;Z)
Hn1(Sn1;Z)n1
GG Hnc (S+n \ Sn1;Z)
yy
where k and k are the Mayer-Vietoris and excision coboundary opera-
tors. Since S+n is an oriented manifold with boundary, we get from Exer-
cise 1.7.20 that
Sn n1(cn1) = S+n \Sn1 n1(cn1) = Sn1 cn1where the sign + appears if Sn1 is oriented as the boundary of S+n and
the sign appears in the other case (a simple computation shows that thesign is in fact (1)n).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
47/236
1.8. Homotopy theorem 35
For n = 1, we have the commutative diagram
H
0
(S
+
1 ;Z
) H
0
(S
1 ;Z
)
GG
O
H
0
(S0;Z
)
0GG
O
H
1
(S1;Z
)O
Z2 1 11 1
GG Z2( 1 1 )
GG Z
Therefore, the generators of H1(S1;Z) are the images by 0 of the functions
1 and 2 defined on S0 by setting
1(1) = 0
1(1) = 1
and 2(1) = 1
2(1) = 0Since
S1
0
2 = (2(1) 2(1)) = 1
we may take vS1 = 02.
Assume now that n > 1 and that we have found vSn1 Hn1(Sn1;Z)
such that
Sn1vSn1 = 1.
Since Sn
n1
vSn1 = (1)n
Sn1
vSn1 = (1)n
we may choose vSn = (1)n
n1vSn1 . The conclusion follows easily by
induction.
Exercise 1.8.6. Assume n 1. Show that if Cn is an open cell ofRn (i.e.an open subset ofRn which is homeomorphic to Bn) then
Hkc (Cn;Z) 0
for k = n and
induces an isomorphism
Hnc (Cn;Z) Z.
In particular, there is a unique class vCn Hnc (Cn;Z) such that
vCn = 1.
Show also that ifCn is an open cell ofRn included in Cn, then the canonical
morphism
Hnc (Cn;Z) H
nc (Cn;Z)
is an isomorphism which sends vCn to vCn .
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
48/236
36 1. Survey of sheaf theory
Solution. Let us first assume that Cn = Bn. We start from the excision
distinguished triangle
Rc(Bn;Z) R(Bn;Z) R(Sn1;Z) +1
For n = 1, we get the exact sequence
H0(Bn;Z) GG
O
H1(S0;Z)0GG
O
H1c(B1;Z)GG
O
0
Z( 11 )
GG Z2( 1 1 )
GG Z GG 0
It follows that a generator of H1c(B1;Z) is vB1 = 0([]) where(1) = 0
(1) = 1
Since B1
vB1 =
B1
0([]) = (1) (1) = 1,
we see that B1 induces the isomorphismB1
: H1c(B1;Z)
Z.
For n 2, we get the isomorphism
Hn1(Sn1;Z)/o
n1GG
O
Hnc (Bn;Z)O
Z1
GG Z
Setting vBn = n1vSn1 , we see thatBn
vBn =
Bn
n1vSn1 =
Sn1
vsn1 = 1
and the conclusion follows.
Assume now Cn is a general open cell ofRn. By functoriality, it is clear
that
H
k
c (Cn;Z
) H
k
c (Bn;Z
) Z if k = n0 otherwiseLet B be an open ball ofRn with center x. By inversion, Rn \ B B \ {x}.Therefore,
Rc(Rn \ B;Z) Rc(B \ {x};Z)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
49/236
1.8. Homotopy theorem 37
From the excision distinguished triangle
Rc(B \ {x};Z) R(B;Z) R({x};Z)+1
it follows that Rc(B \ {x};Z) 0. Hence, Rc(Rn \ B;Z) 0 and
Rc(B;Z) Rc(Rn;Z)
is an isomorphism in the derived category.
It follows that
Hnc (B;Z) Hnc (R
n;Z)
is an isomorphism. If we assume moreover that x Cn and B Cn, we seefrom the commutative diagram
Hnc (B;Z)GG
99
Hnc (Cn;Z)
vvnnnnnn
Hnc (Rn;Z)
that
Hn
c
(Cn;Z) Hn
c
(Rn;Z)
is surjective. Since these groups are both isomorphic to Z, the preceding
morphism is in fact an isomorphism. Using the commutativity of the dia-
gram
Hnc (Bn;Z)/o GG
Bn
77uuuu
uuHnc (R
n;Z)
Rnss
ssss
C
we see that Rn induces an isomorphismRn
: Hnc (Rn;Z) Z.
A similar argument with Bn replaced by Cn allows us to conclude.
Exercise 1.8.7. Let : Cn Cn be a diffeomorphism between two opencells ofRn. Show that
vCn = sgn(J)vCn
where J denotes the Jacobian of .
Solution. It follows from the preceding exercise that
vCn = m vCn .
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
50/236
38 1. Survey of sheaf theory
On one hand, we have
m vCn = mvCn = m.On the other hand, using de Rham cohomology, we get
vCn = sgn(J)
vCn = sgn(J).
The conclusion follows.
Exercise 1.8.8.
(a) Show that the canonical morphism
R{0}(Bn;Z) Rc(Bn;Z)
is an isomorphism.
(b) Deduce from (a) that for any open neighborhoodU ofu in Rn we have
Hk{u}(U;Z) = Z if k = n;0 otherwise.
Show, moreover, that Hn{u}(U;Z) has a unique generatorvu such thatU
vu = 1
(c) Let : U V be a diffeomorphism between open subsets ofRn and
denote J its Jacobian. Show that for any u U, the map
: Hn{(u)}(V;Z) Hn{u}(U;Z)
sends v(u) to sgn(J(u))vu.
Solution. (a) Consider the morphism of distinguished triangles
R{0}(Bn;Z) GG
R(Bn;Z) GG
R(Bn \ {0};Z)+1
GG
Rc(Bn;Z) GG R(Bn;Z) GG R(Sn1;Z)+1
GG
In this diagram, the second vertical arrow is the identity and by the homo-
topy theorem the third vertical arrow is an isomorphism. It follows that the
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
51/236
1.8. Homotopy theorem 39
first vertical arrow is also an isomorphism. Since this arrow is the compo-
sition of the isomorphism
R{0}(Bn;Z) R{0}(Bn;Z)
with the canonical morphism
R{0}(Bn;Z) Rc(Bn;Z),
we get the conclusion.
(b) The cohomology table follows directly from (a) and Exercise 1.8.6.
Let BU U be an open ball with center u. By (a),
Hn{u}(U;Z) Hn{u}(BU;Z) H
nc (BU;Z).
Denote vu the element of Hn{u}(U;Z) corresponding to vBU H
nc (BU;Z).
Since the diagram
Hn{u}(U;Z)GG
Hnc (U;Z)
Hn{u}(BU;Z)GG Hnc (BU;Z)
yy
is commutative, we get U
vu =
BU
vBU = 1.
(c) Thanks to (b), the result may be obtained by working as in the
preceding exercise.
Exercise 1.8.9. Let Bn+1 = {(x0, , xn) Bn+1 : x0 0} and identify
Bn with{(0, x1, , xn) : (x1, , xn) Bn}.
(see Figure 1.8.1). Denote vBn+1\Bn Hn+1c (B
n+1 \ Bn;Z) and vBn
Hnc (Bn;Z) the classes which have an integral equal to 1. Show that
vBn+1\Bn
= n(vBn )
wheren is the coboundary operator associated to the distinguished triangle
Rc(Bn+1 \ Bn;Z) R(B
n+1;Z) Rc(Bn;Z)
+1
Solution. Thanks to the preceding exercise, this follows directly from Exer-
cise 1.7.20.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
52/236
40 1. Survey of sheaf theory
Figure 1.8.1:
x0
(x1, , xn)
Bn
Bn+1
1.9 Cohomology of compact polyhedra
Definition 1.9.1. A finite simplicial complex is a finite set of non-empty
finite sets called simplexes such that if and then . Wecall the finite set S = , the set of vertices of .
The dimension of a simplex is the number dim = # 1.
Simplexes of dimension n N are called n-simplexes. A p-face of an n-simplex is a p-simplex such that . The dimension of the finitesimplicial complex is the number
dim = sup
dim .
The n-skeleton of is the simplicial complex
n = { : dim n}.
The realization of is the compact subspace || of R#S
defined bysetting
|| = { : S R : | 0, supp ,sS
(s) = 1}.
If , we set
|| = { || : supp = }.
Clearly, || || = if and only if = and || =
||.
The data of a finite simplicial complex and an isomorphismh : || X
of topological spaces is a finite triangulation of X. A topological space X
which has a finite triangulation is a compact polyhedron.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
53/236
1.9. Cohomology of compact polyhedra 41
Examples 1.9.2. The set whose elements are
{A}, {B}, {C}, {D}, {A, B}, {A, D}, {B, D}, {B, C}, {D, C}, {A,B,D}
is a finite simplicial complex of dimension 2. It has {A, B, C , D} as set ofvertices and contains
- four 0-simplexes ({A}, {B}, {C}, {D}),
- five 1-simplexes ({A, B}, {A, D}, {B, D}, {B, C}, {D, C}),
- one 2-simplex ({A, B, D}).
A compact polyhedron homeomorphic to || is
A
B
C
D
Compact polyhedra homeomorphic to |1| and |0| are respectively
A
B
C
D
and
A
B
C
D
Definition 1.9.3. Let be a k-simplex of . Two bijections
: {0, , k} , : {0, , k}
have the same sign if the signature of 1 is positive. Clearly, therelation to have the same sign is an equivalence relation on the set of
bijections between {0, , k} and . An equivalence class for this relationis called an orientation of . Of course, a k-simplex of has only two
possible orientations, if o is one of them, we denote o the other. An
oriented k-simplex is a k-simplex of endowed with an orientation. If : {0, , k} is a bijection, we denote by [(0), , (k)] the orientedk-simplex obtained by endowing with the orientation associated to .
A k-cochain of is a map c from the set of oriented k-simplexes of to
Z such that if c(,o) denotes the values of c on the oriented simplex (, o),
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
54/236
42 1. Survey of sheaf theory
we have c(,o) = c(,o). It is clear that k-cochains form a group. Wedenote it by Ck(). We define the differential
dk : Ck() Ck+1()
by setting
(dkc)[x0, ,xk+1] =k+1l=0
(1)lc[x0, ,xl, ,xk+1].One checks easily that the groups Ck() (k 0) together with the differ-
entials dk (k 0) form a complex C() canonically associated with . Wecall it the simplicial cochain complex of .
Lemma 1.9.4. Let = {x0, , xk} be a k-simplex of and let l be the
point of || corresponding to xl. Set
(x0, ,xk)(t1, , tk) = 0 +kl=1
tl(l 0)
and
Jk = {(t1, , tk) R
k
: t1 > 0, , tk > 0,
k
l=1 tl < 1}.Then, Jk is an open cell of R
k and (x0, ,xk ) : Jk || is an homeo-morphism. Moreover, if v(x0, ,xk) is the image of vJk by the isomorphism
1(x0, ,xk)
: Hkc (Jk;Z)
Hkc (||;Z), then we have
v(x0 , ,xk ) = (sign )v(x0, ,xk)
for any permutation of {0, , k}.
Proof. The fact that is an homeomorphism is obvious. Let be a per-
mutation of {0, , k}. Set
= 1(x0 , ,xk )
(x0, ,xk).
Clearly, is the restriction to Jk of the affinity ofRk characterized by
(Pl) = Pl
where P0 = 0, P1 = e1, , Pk = ek. It follows that preserves orreverses the orientation ofRk according to the fact that is even or odd.
Using Exercise 1.8.7, we see that
(vJk ) = (sign )vJk .
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
55/236
1.9. Cohomology of compact polyhedra 43
Hence,
1(x
0, ,x
k
)(vJk ) =
1(x0, ,xk)
(vJk )
= (sign )1(x0, ,xk)(vJk )
and the conclusion follows.
Lemma 1.9.5. Let = {x0, , xk} be a k-simplex of and let =
{x1, , xk}. The distinguished triangle
Rc(||;Z) Rc(|| ||;Z) Rc(|
|;Z)+1
induces a canonical morphism
k1 : Hk1c (||;Z) Hkc (||;Z)
and we have
k1(v[x1, ,xk]) = v[x0, ,xk].
Proof. The existence of the distinguished triangle follows from the fact that
|| (resp. ||) is open (resp. closed) in || ||. Thanks to the morphism
(x0, ,xk) of the preceding lemma, we may assume that || = Jk, |
| ={(t1, , tk) : t1 > 0, , tk > 0,
kl=1 tk = 1}. Then, || |
| appearsas a manifold with boundary and the result follows from Exercises 1.8.9
and 1.7.20 since
1 1 1 11 1 0 01 0 1 0...
......
. . ....
1 0 0 1
> 0.
Proposition 1.9.6. For any k 0, there is a canonical isomorphism be-tween Hk(||;Z) and Hk(C()).
Proof. Let us consider the excision distinguished triangle
Rc(|p| \ |p1|;Z) Rc(|p|;Z) Rc(|p1|;Z)+1
(*)
Since |p| \ |p1| = p \p1 || and || is open in |p| for any p \ p1, we get
Rc(|p| \ |p1|;Z)
p \p1
Rc(||;Z).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
56/236
44 1. Survey of sheaf theory
Using the fact that || is homeomorphic to an open ball ofRp if p \p1, we see that
Hkc (|p| \ |p1|;Z) =
p \p1
Hpc (||;Z) if k = p
0 otherwise.
Using Lemma 1.9.4, we get a canonical isomorphism
Hpc (|p| \ |p1|;Z) Cp().
The long exact sequence of cohomology associated to (*) is
Hkc (|p| \ |p1|;Z)GG Hk(|p|;Z) GG H
k(|p1|;Z) -,*+/.()
GG Hk+1c (|p| \ |p1|;Z)GG Hk+1(|p|;Z) GG H
k+1(|p1|;Z)
For k > p, we get that
Hk(|p|;Z) Hk(|p1|;Z).
By decreasing induction on p, we see that Hk(|p|;Z) Hk(|0|;Z) 0.For k < p 1, we obtain
Hk(|p|;Z) Hk(|p1|;Z).
By increasing induction on p, this gives us the isomorphism
Hk(|p1|;Z) Hk(||;Z).
For k = p 1, we get the exact sequence
0 GG Hp1(|p|;Z) GG Hp1(|p1|;Z) -,*+
/.()GG Hpc(|p| \ |p1|;Z)
GG Hp(|p|;Z) GG Hp(|p1|;Z)
Using the isomorphisms obtained above, we may rewrite this sequence as
0 Hp1(||;Z)p1
Hp1(|p1 |;Z)p1
Cp (||)p
Hp(|p|;Z) 0.
Set p = p p. Clearly, Cokerp1 Coker p1 Hp(|p|;Z) andthrough this isomorphism, the canonical map
(p) : Coker p1 Cp+1(||)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
57/236
1.9. Cohomology of compact polyhedra 45
becomes the map
p : Hp(|p|;Z) Cp+1(||).
It follows that Ker p/ Im p1 is canonically isomorphic to
Ker p Hp(||;Z).
To conclude, it remains to show that p = dp. Let [x0, , xp+1] be anoriented (p + 1)-simplex of and let c Cp() be a simplicial p-cochain.We have to show that
[p(c)][x0, ,xp+1] =
p+1
l=0(1)lc[x0, ,
xl, ,xp+1].
Denote the simplicial complex formed by the finite non-empty subsets of
{x0, , xp+1}.
Using the morphisms of distinguished triangles
Rc(|k| \ |k1|;Z) GG Rc(|k|;Z) GG Rc(|k1|;Z)+1
GG
Rc(|k| \ |k1|;Z) GG
yy
Rc(|k|;Z) GG
yy
Rc(|k1|;Z)+1
GG
yy
One sees that we may assume = . Moreover, we may assume that
c[x0, ,xl, ,xp+1] =
1 if l = 0
0 otherwise
Set = {x0, , xp+1} and = {x1, , xp+1}. Since || || is openin || and || is open in |p1|, we have the morphism of distinguished
triangles
Rc(||;Z) GG
Rc(|| ||;Z) GG
Rc(||;Z)
+1GG
Rc(|| \ |p|;Z) GG Rc(||;Z) GG Rc(|p|;Z)+1
GG
Therefore, the diagram
Hpc(||;Z)
GG
Hp+1c (||;Z)
Hpc(|p| \ |p1|;Z)
Hp(|p|;Z) GG Hp+1c (|| \ |p|;Z)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
58/236
46 1. Survey of sheaf theory
is commutative. Since c is the image of the orientation class v[x1, ,xp] Hpc (|
|;Z) by the canonical morphism
Hpc (||;Z) Hpc (|p| \ |p1|;Z) C
p(),
p(c) is the image of (v[x1, ,xp]) by the canonical morphism
Hp+1c (||;Z) Hp+1c (|| \ |p|;Z) C
p+1().
But Lemma 1.9.5 shows that
(v[x1, ,xp]) v[x0, ,xp]
and the conclusion follows.
Corollary 1.9.7 (Eulers theorem). Let be a finite simplicial complex.
Then, the abelian groups
Hk(||;Z) k 0
are finitely generated and the Euler-Poincare characteristic
(||) = k
(1)k rk Hk(||;Z)
is equal to k
(1)k#(k \ k1).
Proof. The first part follows directly from the fact that
C
k
() Z
#(k\k1 )
is a free abelian group with finite rank.
Denote Zk() and Bk+1() the kernel and the image of the differential
dk : Ck() Ck+1()
and set
Hk() = Zk()/Bk().
From the exact sequences
0 Zk() Ck() Bk+1() 0
0 Bk() Zk() Hk() 0
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
59/236
1.9. Cohomology of compact polyhedra 47
we deduce that
rk Ck() = rk Zk() + rk Bk+1()
and thatrk Zk() = rk Bk() + rk Hk().
It follows that
rk Ck() = rk Bk() + rk Bk+1() + rk Hk()
and, hence, that k
(1)k rk Ck() =k
(1)k rk Hk().
The conclusion follows since
rk Ck() = #(k \ k1)
and
Hk() Hk(||;Z).
Examples 1.9.8.
(a) Consider a simplicial complex with || homeomorphic to
Clearly, has four 0-simplexes, five 1-simplexes and one 2-simplex. Hence,
(||;Z) = 4 5 + 1 = 0.
(b) Consider a simplicial complex with || homeomorphic to
We have three 0-simplexes and three 1-simplexes. Therefore, (||;Z) =3 3 = 0. Note that || S1 and our result is compatible with the factthat
Hk(S1;Z) Z if k = 0, 1
0 otherwise
since dim H0(S1;Z) dimH1(S1;Z) = 1 1 = 0.(c) Similarly, for
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
60/236
48 1. Survey of sheaf theory
we have = 4 6 + 4 = 2. This is compatible with the relation
dimH0(S2;Z) dimH1(S2;Z) + dim H
2(S2;Z) = 1 0 + 1 = 2.
1.10 Cohomology of locally compact spaces
Definition 1.10.1. Let X be a locally compact topological space. We call
the cohomological dimension of the functor
F c(X; F)
the cohomological dimension of X and denote it by dimc X.
Remark 1.10.2. Note that by a well-known result about cohomological
dimensions, dimc X n (n N) if and only if
Hk
c (X; F) 0
for any k > n and any F Shv(X). Note also that contrarily to what
may appear at first glance, cohomological dimension is a local notion. More
precisely, if U is an open covering of X, we have
dimc X = supUU
dimc U.
In particular, dimc U dimc X for any open subspace U of X. Note that
although a similar majoration holds for closed subspaces, it may be false forarbitrary subspaces.
Exercise 1.10.3. Show that the cohomological dimension of an open sub-
space of Rn is equal to n. Deduce from this fact that if U and V are
homeomorphic open subspaces ofRn andRm then n = m.
Solution. Let us prove that dimcRn n. The conclusion will follow sincewe know that
Hnc (B;Z) = Z
for any open ball B ofRn.
Since1
2+
1
arctg x : R ]0, 1[
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
61/236
1.10. Cohomology of locally compact spaces 49
is a homeomorphism, Rn is homeomorphic to ]0, 1[n. Since this last spaceis an open subspace of [0, 1]n, it is sufficient to show that dimc [0, 1]
n n.
We will proceed by induction on n.For n = 1, this follows from Exercise 1.6.7. Assuming the result is true
for n, we prove it for n + 1 by using the isomorphism
R([0, 1]n+1; F) R([0, 1];p(F))
(where p : Rn+1 Rn is a canonical projection) and the fibers formula forRp.
The last part follows from the fact that cohomological dimension is
clearly invariant by homeomorphism.
Remark 1.10.4. It follows at once from the preceding exercise that the
cohomological dimension of a differential manifold is equal to its usual di-
mension.
Exercise 1.10.5. Let be a finite simplicial complex. Show that the
cohomological dimension of || is equal to dim.
Solution. We will proceed by induction on dim . For dim = 0, the result
is obvious. To prove that the result is true for dim = n +1 if it is true for
dim n, it is sufficient to use the excision distinguished triangle
Rc(|n+1| \ |n|; F) Rc(|n+1|; F) Rc(|n|; F)+1
together with the fact that |n+1| \ |n| is a finite union of open cells ofdimension n + 1.
Definition 1.10.6. The reduced cohomology H(X;Z) ofX with coefficient
in Z is defined by setting
Hk(X;Z) =H0(X;Z)/Z if k = 0,
Hk(X;Z) otherwise.
A topological space X is cohomologically locally connected (clc for short) if,
for any x X and any neighborhood U of x, there is a neighborhood V ofx included in U such that all the morphisms
Hk(U;Z) Hk(V;Z)
are zero.
Examples 1.10.7. One checks directly that differential manifolds are clc
spaces. With a little more work, one sees also that the same is true of
compact polyhedra.
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
62/236
50 1. Survey of sheaf theory
Proposition 1.10.8 (Borel-Wilder). AssumeX is a locally compact clc-
space. Then, for any pairK, L of compact subsets of X such that L K,
all the restriction morphismsHk(K;Z) Hk(L;Z)
have finitely generated images.
Proof. We will proceed by increasing induction on k. Denote L the familyof compact subsets L of K which have a compact neighborhood L K
for which the image
rk
LK : Hk
(K;Z) Hk
(L
;Z)
is finitely generated. It is clear that any point x K has neighborhoodin L. Hence, it is sufficient to show that if L1, L2 L then L1 L2 L.Choose compact neighborhoods L1, L
2 of L1, L2 for which r
kL1K
and rkL2Khave finitely generated images. Let L1 , L
2 be compact neighborhoods of
L1, L2 such that L1 L
1
, L2 L
2
. Consider the diagram
Hk(K;Z)
GG
Hk(K;Z) Hk(K;Z)
Hk1(L1 L2;Z)
GG
Hk(L1 L2;Z)
GG Hk(L1;Z) H
k(L2;Z)
Hk1(L1 L2 ;Z)
GG Hk(L1 L2 ;Z)
where the horizontal morphisms come from Mayer-Vietoris sequences and
the vertical ones are restriction maps. We know that Im is finitely gener-
ated. Since
(Im ) Im we see that (Im ) is also finitely generated. Hence, so is Im /(Im Im ). Using the epimorphism
(Im /(Im Im )) Im( )/ Im( )
we see that Im( )/ Im( ) is finitely generated. Since the inductionhypothesis shows that Im is finitely generated, it follows that Im( )
is finitely generated. This shows that L1 L2 L and the conclusion
follows.
Corollary 1.10.9. Assume X is a compact clc space. Then, the abelian
groups
Hk(X;Z)
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
63/236
1.11. Poincare-Verdier duality 51
are finitely generated. If, moreover, X has finite cohomological dimension,
then the Euler-Poincare characteristic
(X) = kZ
rk Hk(X;Z)
is well-defined.
Remark 1.10.10. It follows from the preceding corollary that a compact
differential manifold has a well-defined Euler-Poincare characteristic. We
will study it with more details in Chapter 2.
1.11 Poincare-Verdier duality
Let f : X Y be a continuous map between locally compact spaces.
Definition 1.11.1. A closed subset F of X is f-proper if the map
f|F : F Y
is proper or in other words if F f1(K) is compact for every compact
subset K of Y. Clearly, f-proper subsets of X form a family of supports.Let F be a sheaf on X and let U be an open subset of Y. We set
f!(F)(U) = fproper(f1(U); F).
One checks easily that f!(F) is a sheaf on Y. We call it the direct imagewith proper supports of F by f.
Proposition 1.11.2. The functor
f! : Shv(X) Shv(Y)
is left exact and has a right derived functor
Rf! : D+(Shv(X)) D
+(Shv(Y))
which is computable by means of c-soft resolutions.
Remark 1.11.3. For the canonical map
aX : X {pt}
we see easily that
RaX !(F) Rc(X; F).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
64/236
52 1. Survey of sheaf theory
Proposition 1.11.4 (Fibers formula). For any F D+(Shv(X)), wehave a canonical isomorphism
[Rf!(F)]y Rc(f1(y); F).
Corollary 1.11.5. The cohomological dimension of the functor f! is equal
to
supyY
[dimc f1(y)]
Corollary 1.11.6 (Cartesian square formula). Assume
Yf
GG
X
T
gyy
fGG Z
gyy
is a cartesian square of locally compact spaces. Then, we have the canonical
isomorphisms
g1f! f!g
1 and g1Rf! Rf
!g1
Proposition 1.11.7. Let g : Y Z be another continuous map of locallycompact spaces. Then, there are canonical isomorphisms
(g f)! g! f! and R(g f)! Rg! Rf!.
Remark 1.11.8. Combining the preceding result with Remark 1.11.3, we
see that
Rc(Y; Rf!F) Rc(X; F).
A result which may be seen as a kind of Leray theorem with compact sup-ports.
Theorem 1.11.9 (Poincare-Verdier duality). Assume f! has finite co-
homological dimension (i.e. assume that there is n 0 such that
Hk(Rf!F) = 0
for k > n and any F Shv(X)). Then,
Rf! : D+(Shv(X)) D+(Shv(Y))
has a right adjoint
f! : D+(Shv(Y)) D+(Shv(X)).
7/28/2019 Introduction to Characteristic Classes and Index Theory Jean-pierre Schneiders
65/236
1.11. Poincare-Verdier duality 53
In other words, there is a canonical functorial isomorphism
HomD+(Shv(Y))
(Rf!F, G) Hom D+(Shv(X))(F, f!G).
Moreover, for F D(Shv(X)) and G D+(Shv(Y)), there is a functorialisomorphism
RHom(Rf!F